Make set_recommended_min_free_kbytes() callable from outside
khugepaged.c by removing the static qualifier and adding a
declaration in include/linux/khugepaged.h.

This allows callers that change THP settings to recalculate
watermarks without going through start_stop_khugepaged().

Suggested-by: Lorenzo Stoakes (Oracle) <ljs@kernel.org>
Signed-off-by: Breno Leitao <leitao@debian.org>
---
 include/linux/khugepaged.h | 1 +
 mm/khugepaged.c            | 2 +-
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/include/linux/khugepaged.h b/include/linux/khugepaged.h
index d7a9053ff4fed..76252865fca4b 100644
--- a/include/linux/khugepaged.h
+++ b/include/linux/khugepaged.h
@@ -16,6 +16,7 @@ extern void __khugepaged_exit(struct mm_struct *mm);
 extern void khugepaged_enter_vma(struct vm_area_struct *vma,
 				 vm_flags_t vm_flags);
 extern void khugepaged_min_free_kbytes_update(void);
+extern void set_recommended_min_free_kbytes(void);
 extern bool current_is_khugepaged(void);
 void collapse_pte_mapped_thp(struct mm_struct *mm, unsigned long addr,
 		bool install_pmd);
diff --git a/mm/khugepaged.c b/mm/khugepaged.c
index 1dd3cfca610db..56a41c21b44c9 100644
--- a/mm/khugepaged.c
+++ b/mm/khugepaged.c
@@ -2630,7 +2630,7 @@ static int khugepaged(void *none)
 	return 0;
 }
 
-static void set_recommended_min_free_kbytes(void)
+void set_recommended_min_free_kbytes(void)
 {
 	struct zone *zone;
 	int nr_zones = 0;

-- 
2.47.3

On Thu, Mar 05, 2026 at 06:04:53AM -0800, Breno Leitao wrote:
> Make set_recommended_min_free_kbytes() callable from outside
> khugepaged.c by removing the static qualifier and adding a
> declaration in include/linux/khugepaged.h.
>
> This allows callers that change THP settings to recalculate
> watermarks without going through start_stop_khugepaged().
>
> Suggested-by: Lorenzo Stoakes (Oracle) <ljs@kernel.org>
> Signed-off-by: Breno Leitao <leitao@debian.org>

> ---
>  include/linux/khugepaged.h | 1 +
>  mm/khugepaged.c            | 2 +-
>  2 files changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/include/linux/khugepaged.h b/include/linux/khugepaged.h
> index d7a9053ff4fed..76252865fca4b 100644
> --- a/include/linux/khugepaged.h
> +++ b/include/linux/khugepaged.h
> @@ -16,6 +16,7 @@ extern void __khugepaged_exit(struct mm_struct *mm);
>  extern void khugepaged_enter_vma(struct vm_area_struct *vma,
>  				 vm_flags_t vm_flags);
>  extern void khugepaged_min_free_kbytes_update(void);
> +extern void set_recommended_min_free_kbytes(void);

Please drop the extern, it's a historic artifact that we clean up when we add
new entries (I don't blame you for including it, it's a sort of 'unwritten'
convention :P)

Also, can we put this in mm/internal.h please? As we want to limit who can call
this even in-kernel.

>  extern bool current_is_khugepaged(void);
>  void collapse_pte_mapped_thp(struct mm_struct *mm, unsigned long addr,
>  		bool install_pmd);
> diff --git a/mm/khugepaged.c b/mm/khugepaged.c
> index 1dd3cfca610db..56a41c21b44c9 100644
> --- a/mm/khugepaged.c
> +++ b/mm/khugepaged.c
> @@ -2630,7 +2630,7 @@ static int khugepaged(void *none)
>  	return 0;
>  }
>
> -static void set_recommended_min_free_kbytes(void)
> +void set_recommended_min_free_kbytes(void)
>  {
>  	struct zone *zone;
>  	int nr_zones = 0;
>
> --
> 2.47.3
>

THanks, Lorenzo

On Fri, Mar 06, 2026 at 11:20:57AM +0000, Lorenzo Stoakes (Oracle) wrote:
> >  include/linux/khugepaged.h | 1 +
> >  mm/khugepaged.c            | 2 +-
> >  2 files changed, 2 insertions(+), 1 deletion(-)
> >
> > diff --git a/include/linux/khugepaged.h b/include/linux/khugepaged.h
> > index d7a9053ff4fed..76252865fca4b 100644
> > --- a/include/linux/khugepaged.h
> > +++ b/include/linux/khugepaged.h
> > @@ -16,6 +16,7 @@ extern void __khugepaged_exit(struct mm_struct *mm);
> >  extern void khugepaged_enter_vma(struct vm_area_struct *vma,
> >  				 vm_flags_t vm_flags);
> >  extern void khugepaged_min_free_kbytes_update(void);
> > +extern void set_recommended_min_free_kbytes(void);
>
> Please drop the extern, it's a historic artifact that we clean up when we add
> new entries (I don't blame you for including it, it's a sort of 'unwritten'
> convention :P)

Oh, and with that fixed, feel free to add:

Reviewed-by: Lorenzo Stoakes (Oracle) <ljs@kernel.org>

To this patch on respin.

Cheers, Lorenzo